English

On Learning and Learned Data Representation by Capsule Networks

Computer Vision and Pattern Recognition 2020-06-23 v3

Abstract

In this work, we investigate the following: 1) how the routing affects the CapsNet model fitting; 2) how the representation using capsules helps discover global structures in data distribution, and; 3) how the learned data representation adapts and generalizes to new tasks. Our investigation yielded the results some of which have been mentioned in the original paper of CapsNet, they are: 1) the routing operation determines the certainty with which a layer of capsules pass information to the layer above and the appropriate level of certainty is related to the model fitness; 2) in a designed experiment using data with a known 2D structure, capsule representations enable a more meaningful 2D manifold embedding than neurons do in a standard convolutional neural network (CNN), and; 3) compared with neurons of the standard CNN, capsules of successive layers are less coupled and more adaptive to new data distribution.

Keywords

Cite

@article{arxiv.1810.04041,
  title  = {On Learning and Learned Data Representation by Capsule Networks},
  author = {Ancheng Lin and Jun Li and Zhenyuan Ma},
  journal= {arXiv preprint arXiv:1810.04041},
  year   = {2020}
}
R2 v1 2026-06-23T04:33:35.380Z